This is my final project for the Google Data Analytics Professional Certificate course. Throughout this project, I will be following the data analysis cycle, which are Ask, Prepare, Process, Analyze, Share and Act.

SCENARIO

Assuming the position of a junior data analyst, I am working on the marketing team of Bellabeats, a high-tech manufacturer of health-focused products for women. The company’s offerings include a membership plan, the Bellabeats app, the “Leaf” tracker, the “Time” watch and the “Spring” water bottle, all measuring essential parts of users’ daily activity to help them make wise decisions relating to stress management, sleep, menstrual cycle, among others. Having already positioned itself as a rapidly growing tech-driven wellness company for women, the company seeks to explore new opportunities for growth by analyzing similar smart device users’ fitness data for insights. I have been tasked with delivering these insights.

KEY STAKEHOLDERS

  1. Urška Sršen: Bellabeat’s cofounder and Chief Creative Officer
  2. Sando Mur: Mathematician and Bellabeat’s cofounder, also a key member of the Bellabeat executive team.
  3. Bellabeat Marketing Analytics team +I am working as a junior data analyst under this team

PHASE 1: THE ASK

Sršen asks me to analyze smart device usage data in order to gain insight into how consumers use non-Bellabeat smart devices. She wants me to select one Bellabeat product to apply these insights to in my presentation. The guiding questions are: 1. What are some trends in smart device usage? 2. How could these trends apply to Bellabeat customers? 3. How could these trends help influence Bellabeat marketing strategy?

EXPECTATIONS

  1. A clear summary of the business task
  2. A description of all data sources used
  3. Documentation of any cleaning or manipulation of data
  4. A summary of your analysis
  5. Supporting visualizations and key findings
  6. Your top high-level content recommendations based on your analysis

PHASE 2

Sršen recommends using the FitBit Fitness Tracker Data found on Kaggle. 30 eligible Fitbit users consented to submit their personal tracker data, including minute-level output for physical activity, heart rate, and sleep monitoring. It also includes information about daily activity, steps, and heart rate that can be used to explore users’ habits.

Dataset Limitations

  1. A sample size of 30 may not accurately reflect trends from entire population of smart fitness device users.
  2. The data was collected in 2016 and thus, may not accurately depict current user trends.
  3. Having been collected via a third party survey, we are unsure of data integrity.

Reading Datasets

daily_activity <- read.csv("Datasets/dailyActivity_merged.csv")
sleep_pattern <- read.csv("Datasets/sleepDay_merged.csv")
hourly_calories <- read.csv("Datasets/hourlyCalories_merged.csv")
hourly_intensities <- read.csv("Datasets/hourlyIntensities_merged.csv")
hourly_steps <- read.csv("Datasets/hourlySteps_merged.csv")
hourly_data <- cbind(hourly_intensities,Calories=hourly_calories$Calories)
hourly_data <- cbind(hourly_data,TotalSteps=hourly_steps$StepTotal)
weight= read.csv("Datasets/weightLogInfo_merged.csv")

PHASE 3

#install.packages(c("dbplyr", "tidyverse", "janitor","plotly", "cowplot"))

After loading our data, needed packages and libraries, we now want to see how the data looks like, we are working with the daily_activity dataset first and we will start by checking for null values in each column.

daily_activity %>%
  summarise(across(everything(), ~ sum(is.na(.))))
##   Id ActivityDate TotalSteps TotalDistance TrackerDistance
## 1  0            0          0             0               0
##   LoggedActivitiesDistance VeryActiveDistance ModeratelyActiveDistance
## 1                        0                  0                        0
##   LightActiveDistance SedentaryActiveDistance VeryActiveMinutes
## 1                   0                       0                 0
##   FairlyActiveMinutes LightlyActiveMinutes SedentaryMinutes Calories
## 1                   0                    0                0        0

Also, we want to confirm that indeed 30 users responded to the survey as claimed, and check the structure of the columns in the dataset.

print(paste0("Unique User Id's: ", length(unique(daily_activity$Id))))
## [1] "Unique User Id's: 33"
str(daily_activity)
## 'data.frame':    940 obs. of  15 variables:
##  $ Id                      : num  1.5e+09 1.5e+09 1.5e+09 1.5e+09 1.5e+09 ...
##  $ ActivityDate            : chr  "4/12/2016" "4/13/2016" "4/14/2016" "4/15/2016" ...
##  $ TotalSteps              : int  13162 10735 10460 9762 12669 9705 13019 15506 10544 9819 ...
##  $ TotalDistance           : num  8.5 6.97 6.74 6.28 8.16 ...
##  $ TrackerDistance         : num  8.5 6.97 6.74 6.28 8.16 ...
##  $ LoggedActivitiesDistance: num  0 0 0 0 0 0 0 0 0 0 ...
##  $ VeryActiveDistance      : num  1.88 1.57 2.44 2.14 2.71 ...
##  $ ModeratelyActiveDistance: num  0.55 0.69 0.4 1.26 0.41 ...
##  $ LightActiveDistance     : num  6.06 4.71 3.91 2.83 5.04 ...
##  $ SedentaryActiveDistance : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ VeryActiveMinutes       : int  25 21 30 29 36 38 42 50 28 19 ...
##  $ FairlyActiveMinutes     : int  13 19 11 34 10 20 16 31 12 8 ...
##  $ LightlyActiveMinutes    : int  328 217 181 209 221 164 233 264 205 211 ...
##  $ SedentaryMinutes        : int  728 776 1218 726 773 539 1149 775 818 838 ...
##  $ Calories                : int  1985 1797 1776 1745 1863 1728 1921 2035 1786 1775 ...

Key Observations Thus Far:

  • The daily_activity dataset has 15 columns and 940 rows.
  • There are no null values in any of the columns.
  • Values in all the columns are in the correct format, except for the ActivityDate column, which is formatted as a character (chr) type, instead of Date.
  • 33 users responded to the survey, instead of the expected 30. We will have to investigate why this it.
  • We may want to rename the columns for a cleaner look.

Having made these observations, we will now go ahead and clean the data.

Change ActivityDate column to Date format

We will now clean column names easily with the clean_names function from the janitor package

daily_activity=clean_names(daily_activity)
colnames(daily_activity)
##  [1] "id"                         "activity_date"             
##  [3] "total_steps"                "total_distance"            
##  [5] "tracker_distance"           "logged_activities_distance"
##  [7] "very_active_distance"       "moderately_active_distance"
##  [9] "light_active_distance"      "sedentary_active_distance" 
## [11] "very_active_minutes"        "fairly_active_minutes"     
## [13] "lightly_active_minutes"     "sedentary_minutes"         
## [15] "calories"

We are done with the preliminary data cleaning. We will now go ahead and manipulate the data to fine-tune it even more.

Extract days from the dates

day <- wday(daily_activity$activity_date, label = TRUE, abbr = FALSE)
daily_activity <- daily_activity %>%
  mutate(day)
str(daily_activity)
## 'data.frame':    940 obs. of  16 variables:
##  $ id                        : num  1.5e+09 1.5e+09 1.5e+09 1.5e+09 1.5e+09 ...
##  $ activity_date             : Date, format: "2016-04-12" "2016-04-13" ...
##  $ total_steps               : int  13162 10735 10460 9762 12669 9705 13019 15506 10544 9819 ...
##  $ total_distance            : num  8.5 6.97 6.74 6.28 8.16 ...
##  $ tracker_distance          : num  8.5 6.97 6.74 6.28 8.16 ...
##  $ logged_activities_distance: num  0 0 0 0 0 0 0 0 0 0 ...
##  $ very_active_distance      : num  1.88 1.57 2.44 2.14 2.71 ...
##  $ moderately_active_distance: num  0.55 0.69 0.4 1.26 0.41 ...
##  $ light_active_distance     : num  6.06 4.71 3.91 2.83 5.04 ...
##  $ sedentary_active_distance : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ very_active_minutes       : int  25 21 30 29 36 38 42 50 28 19 ...
##  $ fairly_active_minutes     : int  13 19 11 34 10 20 16 31 12 8 ...
##  $ lightly_active_minutes    : int  328 217 181 209 221 164 233 264 205 211 ...
##  $ sedentary_minutes         : int  728 776 1218 726 773 539 1149 775 818 838 ...
##  $ calories                  : int  1985 1797 1776 1745 1863 1728 1921 2035 1786 1775 ...
##  $ day                       : Ord.factor w/ 7 levels "Sunday"<"Monday"<..: 3 4 5 6 7 1 2 3 4 5 ...

We can see that the data now has 16 columns, this is because I have added a column for the particular day of the week each date was. But that gives us another problem; the “day” column is the last of all the columns, but ideally we would want it to be next to the “activity_date” column for better understanding.

Relocate the “day” column

daily_activity <- daily_activity %>%
  relocate(day, .after = activity_date)

Create a new totals column for all activity minutes

daily_activity <- daily_activity %>%
  mutate( total_logged_mins = fairly_active_minutes+very_active_minutes+sedentary_minutes+lightly_active_minutes)

Create a column for total hours

daily_activity['total_logged_hours'] <- round(daily_activity$total_logged_mins/60)

daily_activity <- daily_activity %>%
  relocate(total_logged_mins, .before = calories) %>%
  relocate(total_logged_hours, .before = calories)
head(daily_activity)
##           id activity_date       day total_steps total_distance
## 1 1503960366    2016-04-12   Tuesday       13162           8.50
## 2 1503960366    2016-04-13 Wednesday       10735           6.97
## 3 1503960366    2016-04-14  Thursday       10460           6.74
## 4 1503960366    2016-04-15    Friday        9762           6.28
## 5 1503960366    2016-04-16  Saturday       12669           8.16
## 6 1503960366    2016-04-17    Sunday        9705           6.48
##   tracker_distance logged_activities_distance very_active_distance
## 1             8.50                          0                 1.88
## 2             6.97                          0                 1.57
## 3             6.74                          0                 2.44
## 4             6.28                          0                 2.14
## 5             8.16                          0                 2.71
## 6             6.48                          0                 3.19
##   moderately_active_distance light_active_distance sedentary_active_distance
## 1                       0.55                  6.06                         0
## 2                       0.69                  4.71                         0
## 3                       0.40                  3.91                         0
## 4                       1.26                  2.83                         0
## 5                       0.41                  5.04                         0
## 6                       0.78                  2.51                         0
##   very_active_minutes fairly_active_minutes lightly_active_minutes
## 1                  25                    13                    328
## 2                  21                    19                    217
## 3                  30                    11                    181
## 4                  29                    34                    209
## 5                  36                    10                    221
## 6                  38                    20                    164
##   sedentary_minutes total_logged_mins total_logged_hours calories
## 1               728              1094                 18     1985
## 2               776              1033                 17     1797
## 3              1218              1440                 24     1776
## 4               726               998                 17     1745
## 5               773              1040                 17     1863
## 6               539               761                 13     1728

Data Cleaning and Manipulation on the daily_activity dataset is done. We will now turn our attention to the sleep_pattern and hourly_data datasets.

Convert the SleepDay column to Date format, Clean column names, check for null values, as well as number of unique users to log sleep data and hourly data. There are null values in the sleep_pattern dataset: FALSE

sleep_pattern$SleepDay <- mdy(sleep_pattern$SleepDay)
sleep_pattern = clean_names(sleep_pattern)
hourly_data = clean_names(hourly_data)

print(paste0("Unique User Id's in Sleep Dataset: ", length(unique(sleep_pattern$id))))
## [1] "Unique User Id's in Sleep Dataset: 24"
print(paste0("Unique User Id's in Hourly Dataset: ", length(unique(hourly_data$id))))
## [1] "Unique User Id's in Hourly Dataset: 33"

We find that the hourly dataset captures the data of all 33 users, however, only 24 users logged their sleep data. This means 9 users did not log their sleep data.

We will now see what the hourly data looks like

head(hourly_data)
##           id  activity_hour total_intensity average_intensity calories
## 1 1503960366 4/12/2016 0:00              20          0.333333       81
## 2 1503960366 4/12/2016 1:00               8          0.133333       61
## 3 1503960366 4/12/2016 2:00               7          0.116667       59
## 4 1503960366 4/12/2016 3:00               0          0.000000       47
## 5 1503960366 4/12/2016 4:00               0          0.000000       48
## 6 1503960366 4/12/2016 5:00               0          0.000000       48
##   total_steps
## 1         373
## 2         160
## 3         151
## 4           0
## 5           0
## 6           0

Split hourly datetime

hourly_data <- hourly_data %>%
  separate(activity_hour, into = c("date", "hour"), sep = " ", remove = TRUE)%>%
  separate(hour, into = c("hour_of_the_day", "ddd"), sep = ":", remove = TRUE)%>%
  select(-c(ddd))
hourly_data$hour_of_the_day <-as.integer(hourly_data$hour_of_the_day)

Data cleaning and manipulation is now complete for all 3 datasets. We will go ahead and analyze our data.

*PHASE 4

Firstly, we will find the percentage of minutes spent on each activity type

sedentary_percentage <- round(sum(daily_activity$sedentary_minutes)/sum(daily_activity$total_logged_mins)*100,digits=2)
lightly_active_perc <- round(sum(daily_activity$lightly_active_minutes)/sum(daily_activity$total_logged_mins)*100, digits=2)
fairly_active_perc <- round(sum(daily_activity$fairly_active_minutes)/sum(daily_activity$total_logged_mins)*100, digits=2)
very_active_perc <- 
round(sum(daily_activity$very_active_minutes)/sum(daily_activity$total_logged_mins)*100, digits=2)
plot_data <- data.frame(intensityLevel = c(very_active_perc,  lightly_active_perc, fairly_active_perc,sedentary_percentage),
intensityLabel = c("Very Active Minutes", "Lightly Active Minutes","Fairly Active Minutes",  "Sedentary Minutes"))

plot_ly(plot_data, labels = plot_data$intensityLabel, values = plot_data$intensityLevel, type = 'pie',textposition = 'inside', marker=list(colors=c("#93DC5C","#32CD80","#DBF3C9","#32CD32"),line = list(color = 'black', width = 1)), textinfo="percent",hovertext= plot_data$intensityLabel, textfont=list(size=20, color="black"), hole=0.4) %>%
  layout(title = 'Percentage of Total Time Spent on Each Activity Type')

Findings:

  • Too much time - as much as 81.3% of users’ logged time - is spent on sedentary activity.
  • Despite FitBit device being a fitness tracker, it is uncertain that it is actually encouraging users to be more active as users fairly and very active minutes make up only 2.85% of the time.

We will now analyze the relationship between distance covered, total steps taken and calories burned

min_date = min(daily_activity$activity_date)
max_date = max(daily_activity$activity_date)
 c = ggplot(data=daily_activity)+
  geom_point(mapping=aes(x=total_distance,y=calories,color=total_steps))+
  geom_smooth(mapping=aes(x=total_distance,y=calories))+
  labs(x="Distance Covered (in km)", y="Calories Burned", title="Correlation Between Distance Covered, Total Steps and Calories", color="Totals Steps",
       caption=paste0("33 Users Data from: ", min_date, " to ", max_date))+
  scale_color_continuous(type="viridis")
 ggplotly(c)

Findings:

  • As expected, there is a positive correlation between total distance covered and calories, as well as between total steps taken and calories.
  • There are not too many outliers; only two outright outliers.
    • One is at 15 km Distance with 4900 calories burned. This could be due to user being very active for most parts of the distance covered.
    • The other is at 28.03 km with 2690 calories burned. This could be due to user being only lightly active across the distance.

We will now go ahead and analyze the daily step totals.

avg_steps_per_day <- daily_activity %>%
  select(total_steps,day)%>%
  group_by(day)%>%
  summarize(avg_steps=mean(total_steps))

c= ggplot(data=avg_steps_per_day)+
  geom_col(mapping=aes(x=day, y=avg_steps), fill = "lightblue")+
  labs(title="Average Steps Taken Across The Week", x="Day Of The Week", y = "Average Steps By All Users")+
  theme_dark()+
  theme(plot.title = element_text(hjust = 0.5))
ggplotly(c)

Key Notes:

  • According to the Center for Disease Control and Prevention, it is recommended for an adult to take at least 10,000 steps per pay.
  • For all the combined data, however, users did not average over 10,000 steps on any day, which is a disturbing statistic.

This leads us to wonder whether users ever crossed the 10,000 step mark.

filt <-daily_activity%>%
  select(total_steps)%>%
  filter(total_steps >= 10000)
perc_over_10000 = round((nrow(filt)/nrow(daily_activity))*100, digits = 1)
perc_over_10000
## [1] 32.2
  • We find that users crossed the 10,000 recommended step total just 32.2% of the time.
  • This may be because most of them might not know that it is actually recommended to make at least 10,000 steps per day for optimum fitness.

We will now move to analyze the sleep data.

day_of_sleep <- wday(sleep_pattern$sleep_day, label = TRUE, abbr = FALSE)
sleep_pattern <- sleep_pattern %>%
  mutate(day_of_sleep)

avg_sleeptime_bedtime <- sleep_pattern %>%
  select(total_time_in_bed, total_minutes_asleep, day_of_sleep)%>%
  group_by(day_of_sleep)%>%
  summarize(avg_bedtime=mean(total_time_in_bed),avg_sleeptime = mean(total_minutes_asleep), Difference = avg_bedtime - avg_sleeptime)


data <- data.frame(days = avg_sleeptime_bedtime$day_of_sleep, average_bedtime = avg_sleeptime_bedtime$avg_bedtime, average_sleeptime = avg_sleeptime_bedtime$avg_sleeptime)

fig <- plot_ly(data, x = ~days, y = ~average_bedtime, type = 'bar', name = 'Average BedTime')
fig <- fig %>% add_trace(y = ~average_sleeptime, name = 'Average SleepTime')%>%
  layout(yaxis = list(title = 'Duration (Minutes)'), title="Average Bedtime Vs. Average Sleep Time", barmode = 'group')%>%
  layout(xaxis = list(title = 'Days of the Week'))
ggplotly(fig)

The grouped bar chart above shows the difference between the average times spent in bed and the average sleep times for all 24 users who recorded their sleep data. To be more blunt, let us check the mean difference between the two.

filt_sleeptime <-sleep_pattern%>%
  select(total_minutes_asleep)%>%
  filter(total_minutes_asleep/60 >= 7)
perc_over_7 = round((nrow(filt_sleeptime)/nrow(sleep_pattern))*100)

"Bed Times Summary"
## [1] "Bed Times Summary"
summary(sleep_pattern$total_time_in_bed)
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    61.0   403.0   463.0   458.6   526.0   961.0
"Sleep Times Summary"
## [1] "Sleep Times Summary"
summary(sleep_pattern$total_minutes_asleep)
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    58.0   361.0   433.0   419.5   490.0   796.0
tt_avg_bedtime = mean(sleep_pattern$total_time_in_bed)
tt_avg_sleeptime = mean(sleep_pattern$total_minutes_asleep)
paste(paste0("Mean Difference: ", tt_avg_bedtime-tt_avg_sleeptime))
## [1] "Mean Difference: 39.1719128329298"
paste(paste0("Average Total Hours of Sleep: ", round(tt_avg_sleeptime/60, digits=2)))
## [1] "Average Total Hours of Sleep: 6.99"
paste(paste0("Percentage of Sleep Times Over 7 hours: ", perc_over_7))
## [1] "Percentage of Sleep Times Over 7 hours: 56"
  • On average, users stay in bed for 39 minutes before falling asleep.
  • Users sleep for about 7 hours on average.
  • 56% (i.e. 231) out of the total 413 recorded sleep times were over the recommended minimum 7 hours of sleep.
plot1 <- ggplot(data=hourly_data)+
  geom_col(mapping=aes(x=hour_of_the_day, y=total_steps,fill=total_intensity))+
  theme(axis.text.x = element_text(angle = 45))+
  labs(title="Total Steps per Hour", x="Hour of the Day", y="Total Steps Taken")+
  theme(legend.position = "bottom")+
  theme(plot.title = element_text(hjust = 0.5))
plot2 <- ggplot(data=hourly_data)+
  geom_col(mapping=aes(x=hour_of_the_day, y=calories))+
  theme(axis.text.x = element_text(angle = -45))+
  labs(title="Calories Burned Each Hour", x="Hour of the Day", y="Calories Burned")+
  theme_grey()+
  theme(plot.title = element_text(hjust = 0.5))+
  scale_color_viridis_c()
plot_grid(plot1, plot2, labels = "AUTO")

The bar chart for total steps taken per hour, is, as expected, almost synonymous to the bar chart for calories burned per hour. Users’ total steps taken and calories burned peaked at 6pm.

We will now check how users record their weight from the “weight” dataset

head(weight)
##           Id            Date WeightKg WeightPounds Fat   BMI IsManualReport
## 1 1503960366  5/2/2016 23:59     52.6     115.9631  22 22.65           TRUE
## 2 1503960366  5/3/2016 23:59     52.6     115.9631  NA 22.65           TRUE
## 3 1927972279  4/13/2016 1:08    133.5     294.3171  NA 47.54          FALSE
## 4 2873212765 4/21/2016 23:59     56.7     125.0021  NA 21.45           TRUE
## 5 2873212765 5/12/2016 23:59     57.3     126.3249  NA 21.69           TRUE
## 6 4319703577 4/17/2016 23:59     72.4     159.6147  25 27.45           TRUE
##      LogId
## 1 1.46e+12
## 2 1.46e+12
## 3 1.46e+12
## 4 1.46e+12
## 5 1.46e+12
## 6 1.46e+12
weight=clean_names(weight)
total_num_records = count(weight)
total_num_users = length(unique(weight$id))

c=  ggplot(data=weight) +
  geom_bar(mapping=aes(x = weight$is_manual_report),fill="brown")+
  labs(x = "The Weight Was Manually Recorded", y = "Number of Records", title = "Manually Recorded Vs. Auto Recorded Weight Data", caption = paste0("Weight Data Includes ",total_num_records," Total Records From ",total_num_users," Users"))+
  theme_grey()+
   theme(plot.title = element_text(hjust = 0.5))
ggplotly(c)
bmi_categories <- weight %>%
  select(weight_pounds, id, bmi) %>%
  group_by(id) %>%
  filter(row_number()==1) %>%
  mutate(weight_category=case_when(
    bmi < 18.5 ~ "Underweight", 
    bmi >= 18.5 & bmi <25 ~ "Healthy", 
    bmi >= 25 & bmi < 30 ~ "Overweight", 
    bmi >= 30 ~ "Obese")
    )

percentages=bmi_categories%>%
  group_by(weight_category)%>%
  summarize(category_totals=n())%>%
  mutate(category_percentage=category_totals/sum(category_totals)*100)

c <- plot_ly(percentages, labels = percentages$weight_category, values = percentages$category_percentage, type = 'pie',textposition = 'inside', marker=list(colors=c("#18F300","#ff0000","#055C1A"),line = list(color = 'black', width = 1)), textinfo="percent",hovertext= percentages$category_totals, textfont=list(size=20, color="white"), pull=0.015) %>%
  layout(title = 'Weight Categories Per BMI Reading')
ggplotly(c)

The following is the CDC’s weight classification per BMI:

  • If your BMI is less than 18.5, it falls within the underweight range.
  • If your BMI is 18.5 to <25, it falls within the healthy weight range.
  • If your BMI is 25.0 to <30, it falls within the overweight range.
  • If your BMI is 30.0 or higher, it falls within the obesity range.

Key Note:

  • One 3 out of the 8 users are in the “healthy” weight range.

```

PHASE 6: ACT

Recommendations:

  1. For more comprehensive analysis, Bellabeat should gather a larger sample of smart fitness device users’ data, which should also be more recent. However, the following are my recommendations from the FitBit tracker data:

  2. Bellabeat’s primary objective should revolve around reducing users’ sedentary activity. Achieving this goal entails integrating a smart algorithm into the “Leaf” tracker, prompting it to emit beeps or the Bellabeat App to send timely alerts whenever a prolonged sedentary period is detected. By going beyond mere data collection, this proactive approach ensures that Bellabeat’s products genuinely enhance users’ fitness by actively encouraging them to stay active and engaged.

  3. The Bellabeat Marketing Team should conduct research on recommended fitness routine goals, including guidelines from reputable organizations like the Center for Disease Control and Prevention (CDC). By disseminating such valuable information, such as the CDC’s recommended minimum steps per day, users can be reminded of setting appropriate daily activity goals for themselves. Additionally, this data empowers users to identify any unusual trends in their activity patterns and take prompt action towards maintaining a healthy and active lifestyle. It would be good to consider including these goals in the company’s newsletter or as brief “Did You Know” messages to users.

  4. To improve data collection and offer meaningful insights to its users, Bellabeat needs to tackle the problem of inadequate sleep and weight data records. Approximately 37.5% (9 out of 33) of users did not input their sleep data, while only 8 out of the total 33 users recorded their weight data. Such gaps can impede their capacity to make well-informed decisions regarding their health. To tackle this problem, Bellabeat should consider sending reminders to users, encouraging them to keep the “Leaf” tracker device on before going to bed. By doing so, users can generate essential sleep data, enabling them to gain valuable insights into their sleep patterns and make healthier choices. +For instance, our analysis revealed that the average sleep latency among users was 39 minutes. Comparing this with Healthline’s recommended sleep latency of 10 to 20 minutes for adults, users with consistently higher sleep latency times could be alerted to take action. Moreover, a study demonstrated that sleep latency of more than 30 minutes can lead to decreased sleep quality. Recording as much sleep data as posisible would enable users to quickly identify if their sleep metrics fall short of the recommended benchmarks and make timely decisions to improve their sleep health.

  5. Given the apparent user preference for manual weight data entry over automated recording, Bellabeat should contemplate integrating an algorithm enabling users to input their daily weight data. However, it’s important to note that more data is required to confirm the viability of this approach. With only 8 users currently recording their sleep data, it is challenging to draw definitive conclusions from the data.

REFERENCES